Ted Unangst <t...@tedunangst.com> wrote: > On 2020-07-08, Theo de Raadt wrote: > > I think we need something like this. > > > > Documenting it will be a challenge. > > > > I really don't like the name as is too generic, when the control is only > > for a narrow set of "current time" system calls. > > I thought I'd start with something, but lots of questions. > > Should it be per wrapper?
Perhaps not named per syscall, but named per grouping. This causes direct syscalls for time retrieval. Can you find a name for that? > I know in the past we've had some similar > conversations about eliminating syscalls (open/openat) I don't think we'll be doing that, because it doesn't really make the world any better and, as is evident with time handling mangles ktrace too much. > Initial thought is it's easier to make one button, and then document > it in ktrace perhaps? Probably in kdump also. Hang on. Do people ever want to force time calls, outside of ktrace? I doubt it. Should ktrace maybe have a flag, similar to -B with LD_BIND_NOW, which sets the new environment variable? Maybe -T? The problem smells very similar to the root cause for LD_BIND_NOW setting..
